Been running the RTM for about a week now. And I like it a lot. Sure, I wish Vista users would get this for cheap, but despite that, it really is nice. The new taskbar takes some getting used to, but it has some great features (Having shortcut controls in the preview has a ton of potential). It feels more responsive and polished, even on a netbook that Vista would choke on.
Much fewer hiccups compared to Vista (I can switch around between Visual Studio, Office, Eclipse, etc with no delays). There are lots of surprises. Try the math panel (best with a tablet, but mouse is okay), you can burn ISO images, it has basic color calibration and finally, the taskbar icons have been tamed. Aero Peek is nice and Aero overall takes less resources. It works fine even with Intel integrated graphics.
All in all, it really sets Windows back on good ground. It's the first version of Windows in quite some time that I really like running. Vista wasn't as bad as everybody said, and I'm sure Windows 7 won't be as good as some will hope, but for now, I am happy to run it.
This article was pretty light on real data, but one thing that struck me is that they benchmarked the 32bit version. Well, why you would run the 32bit edition unless you had to, I don't know. I'd like to see some benchmarks with XP/Vista/7 32bit and Vista/7 64bit and see those outcomes.
The numbers for Office show very little difference in performance, so I can't help but feel that the responsiveness gains will make 7 more tolerable to use, and the UI improvements seem to be genuinely interesting and useful. While Vista has been fine for me (64bit version), I expect 7 will be better tuned.
Speaking as a CS postdoc working on a National Library Training fellowship, I have to stay the best bet is to start small and start fresh. There is a ton of praise about VistA, the VA system, but it is built on old storage technology and may not support the kind of analytics needed for quality improvement. Also, there is a lot of information captured in notes, which is free text and not really helpful in terms of decision support and continuity of care.
I think a good place to start would be a couple of things. First, a identity system that links patients at various providers. There wouldn't be one global id, it's would be set of identities tied to a provider. Think a medical OpenID, and that's a start. A ID that is tied to a provider, easy to revoke, and so on.
I think a national medical ID program would be just fine. As long as it stayed optional (you don't have to provide it, but you may need to pay upfront or provide a payment option).
The second is a continuity of care document. A interchange to track the current problem list, current medications, current allergies and list of notable procedures. You don't need everything at once, just the places to ask.
As for security, I think that patients don't mind if their doctors see their record. The system has to be useful first, and making it too secure will make it too hard to use. Also, transparency is a huge issue. You can really build trust if patients can see what is in their record and who accessed it. Health networks really need to have a page for patients to log into and see what's going on.
Also, the legal protections matter, but they have to be useful. HIPPA in its current form has many problems.
Have to say, this is a critical point. While unions have a lot of context with them, they had consistently done one thing well: Improving working conditions for their members. If they make highly repetitive work doable for many, I can't help but imagine that it would help IT a ton.
Developers, I see that as being less of a issue, but for daily operational management, it seems pretty clear the workforce need protection. For one, a union could enable the classic "You are fired for doing nothing" when it is really "I am making sure nothing happens" and so on.
Given the state of broadcast television, I can't say blacking some people out wouldn't do them a favor. Okay, you need to get a convertor box and you may have to wait to get one, but if we encourage people in the meantime to read a book, go to the library, use the computer there and read the news and so on, that's bad? Really?
I mean, I'm scared that people think that TV is that much of a requirement. Local news is nice and all, sure, but you can make do.
"Some vaccines are unneeded. Chicken pox, for example..."
Well, until there is science noting that the risks of the vaccine are higher than the risks of chicken pox infections (which can lead to serious complications and deaths in children and adults), there are good reason to vaccinate. The severity of the disease is not the issue. The issue is the vaccine simulating an effective immune response. The CDC guidelines are taken very, very seriously. So, again, look at the science they do before deciding what is needed or not.
But I'd bet the small business owner is a hell of a lot happier following their dreams on their terms. The crazy thing about this society is that we need people to make our sandwiches, cut our meat and so on, but take pride in belitting them. Not everybody can run a business. No everybody has the chances others get. Others may care about other things. There are other ways to contribute to society that aren't driven by the bottom line. Most SBOs I know just love what they do and enjoy it. But, you know what? It's not about the time you put in. Working eighty hours on your business can mean you are super motivated, or you don't have anything else to do.
The small business owner has a chance for promotion, for success, to make a difference, for all that hard work to make a difference. Too many in this country work hard to survive, not thrive. They work 50 hours or more to get by. The working class may not be "the hardest working", but they are the ones who would hardest for the least in return. Maybe things would be a bit better if the SBOs were more humble and took pride in provide a living for others and less in how much they cleared this year.
Firstly, I feel that people are often looking at scooters and low-displacement motorbikes. For urban drivers, we are talking 100 mpg from a vehicle that is in the 2000-3000K range. Oh, and don't forget the people on bicycles. on
Yes, bikes are more risky, but those risks are:
The driver. But, stupid drivers can get hurt in any vehicle. Cars. I got wrecked because a car decided it was okay to head into my lane. The bike stood up, I ran out of road and that's that. The guy didn't even notice.
As more and more people get on two wheels, I just ask this of car drivers:
Pay attention. Get off the phone. Don't fuss around. Start Seeing Motorcycles. Look for us. We are there. We have a right to be on the road too, and a lot of us are really okay.
As for motorcyclists, our part of the bargain:
Know your limits, practice your skills and wear your gear. Take it easy. If you need a race, find a track.
Harley is studied everywhere in marketing. I mean, how many companies have people tattoo their logo on themselves. Wow.
Yes, there are Harley's with fuel injection. This happened a few years ago, and yes, they have a engine that doesn't totally suck. But their bikes are very costly for what you get. I mean, the cheapest is about 8400. Hint. Do not buy that bike. There are way, way better bikes for 2000-3000 less. Hell, you get more for a BMW at that price.
Scanning through the highly moderated article, I didn't see this point raised, which seems critical. The doctors that would be required to do this research to provide doping and drugs are bound by a ethical code of conduct, and this definitely forbids giving people medications with very serious risks to somebody that is healthy.
Some have expression the opinion of "let them be lab rats." Well, I doubt that any good doctors would touch that ethical quagmire with a ten foot pole. I mean, the notion of informed consent alone is a huge red flag. Can you say that somebody that is willing to risk their health and life to potentially (nothing is certain) perform a bit better is truly of a state of mind to consent? And what doctor would do such a thing. Would you want a doctor that doesn't care about your health, but just want they can get away with?
And the ethics are there for a reason. Bioethics exists for a reason. Part of the Human Genome Project was funding to examine carefully and intensively the impacts of the project, how the data should be used, the impacts on society. In fact, there was discuss if the project should be allowed at all.
Of course, things can improve in testing. What I think needs to happen (and has, in the NFL, for example), is that more athletes need to come out and say that they want to test their limits without risking their health. That at the end of the day, they don't want to dope. That they are not willing to do anything to win. They want to test themselves on their terms, not as some "mad scientist" experimental rat.
And the sport fan needs to learn that the best part of sports is not winning or losing, it is the pursuit of excellence and being tested and retested time and time again.
Corticosteroids (the kind used to treat psoriasis) are a completely different animal from anabolic steroids when it comes to human performance. Yes, they share some chemical commonalities, but croticosteroids don't raise testoterone levels (anabolic steroids do). Also, an barbiturates probably aren't considered performance enhancing either.
Of course, all of the above have side effects (I'm sorry your dad doesn't respond to other anti-seizure meds, the barbs suck). But the reasons for this is that those effects are not as bad as the the problems they address.
When I read this, it was more evidence that some in Google want to be "the internet". For very, very many, they are the gatekeepers to the web. Now, with this, they can be the hosts as well. If Google hosts a ton of applications, you can see a future in which somebody is surfing from site to site but never really leaving Google.
So, what's to stop the extingush part? Letting Google say, hey, host here and we will make sure that your search results don't "change", or you get better results, etc. Given the demands on growth and the need for Google to justify it's P/E, it is starting to look for more ways to make money. And Google is large enough to become the "Microsoft" or "IBM" of web applications if it really wants too, and why wouldn't it?
Of course, I don't buy the "Google isn't evil, that's why" prima facie argument.
Yea, it's going to be tricky. Big issue. For the longest time, we got a huge boost from Moore's Law in single-threaded performance, and for desktop OSes (esp. XP, Vista), they are tuned and around interactive, single foreground program workloads. But that's over. And Vista does add IO and memory demands (search, indexing, UI experience, etc.), and some in the market (and MS, of course) want that.
Add in memory hierarchies from hell (L3, it's back, it's crazy) and it does get to the point where you have to think about stuffing as much stuff as you might need into memory, because going to disk just kills you, because going to ram to cache is bad enough.
Well, just to be clear, I just have 4G right now on my workstation and it's more than fine. Even 2G is no problem (older laptop). But, yes, I stand by my statement that for my work, I'd rather have lots of slower memory than tons of fast memory. YMMV of course.
I've been running Vista for about 16 months now. (Got it from MSDN, etc.). I ran both 32bit and 64bit editions, on two kinds of hardware. It's been a bumpy ride sometimes.
If at all possible, use the x64 edition. Yes, some OEM make this a pain, but try. Given this, my next suggestion seems obvious. Get more memory. The more, the better. I'd rather have 8G of DDR2-533 than 2G of DDR3 uber-awesome overclocked OMGBBQ ram, because caching works. If on an Intel integrated graphics, turn down some of the Aero stuff. Duh. If possible, just buy a cheap 8400GS, because even that will help.
SP1 helps. Some things are faster. Of course, I'm not seeing some of the problems others are. I recently shipped some ISOs over from machine to my server (Win 2008), and it just flew. Got about 600Mb over a 1Gb switched link.
All in all. Not one blue screen on the desktop, a couple on the laptop due a older bluetooth driver. Things seem plenty responsive and fast, but there is a breaking in period. Sure, it isn't "awesome", but ME it sure is not.
Be patient! Indexing and the prefetching stuff takes time, but it does work. I use Outlook (okay, I know, I know) a lot, and it fired up faster and faster after the first day or two.
Nothing official, but I hear that the MinWin idea was based a lot on the Singularity project. The link has some papers, and there is some really interesting ideas there in terms of fast process isolation, etc.
Sure, a microkernel is nothing new, but there are new ways of making a microkernel.
Firstly, I think your conclusion that "Java is much faster than C#" is very difficult to support. Of course, there are lies, damn lies, statistics and benchmarks, but the overall picture I see is that the performance of C# is in fact quite competitive with Java in terms of runtime performance.
As for some of the decisions, they are not just about pure speed. There are more considerations.
Value types: This allows the choice of the best semantic match is. Also, this does help with native interoperability, as in some cases, you can model a C struct in C# and pass it back and forth with a lower cost. Also, the complexity cost you note is small, as even Java still has value types. If you have to deal with both, then adding in structs isn't that costly, from a language and runtime view.
JIT-only: The ability to pre-compile to native code helps with native interoperability and performance. It can affect startup time, but there are techniques to deal with this a bit. I think MS thought the added complexity of interpretation+hotspot compliation wasn't worth it for them. It was a tradeoff, but I don't see it as a drastic one. Also, C# can still code-pitch if needed. It is probably harder.
Real Generics: To be honest, I can't say that Java Generics are all that great, from a programming model. Sure, they don't technically cost anything, because they go away. Which causes real problems and limitations.
For example, C# 3.0 can do a ton of type-inference and more functional-like program as the type information
of a delegate (or lambda expression) is kept and passed. Also, I'd like to see the reference for the CLR only inlining one deep.
Embedded Native Code: Well, faster access to native code can make a huge improvement in performance, and this is a real problem with the Java runtime. JNI has some real issues in terms of usability from a programming standpoint. The CLR makes it must easier to call into a native function if you need to, and that's often a really useful point. Not just for Windows, but for OS/X, for Linux, etc.
The GC: Sure, it is more complex, but there isn't much evidence that it is slower because of it. Again, if fast native interoperability is a goal, then this may be a cost, but it may be a good one.
Yes, the CLR is more complex. That is because it can actually model more aspects of an executable. It has pointers. It has "unsafe" code. You can compile any C/C++ program that compiles in Visual Studio 2005/2008 to the CLR with a few switches. In some cases, the performance hit is surprisingly small versus the native code.
With the Phoenix project, the.Net CLR has access to the same optimizations that the native C/C++ compiler has.
All in all, the CLR, C# do represent some interesting ideas and is a very valid platform. The MS equals sucks argument just doesn't hold here. But, it's just an opinion.
I gone through the initial process with Google twice, with the same outcome. It seems to me they need to improve their HR process, as I've gone through a phone interview, but then told I wasn't a good fit.
If you look at my background and resume, I think you would concur that the positions I was interviewed for weren't a good fit, but because it was Google, I gave it a shot. But, fool me once, fool me twice and all that. If they call again, I'll let them know how I feel about the whole process.
If the FSF actually tried to do this, it'd be great for Microsoft. It would basically allow Microsoft to say that Linux can't be taken seriously by any business, because the FSF will basically revoke your "right to use" Linux if they don't like how you do business (do you have software patents? No Linux for you).
It wouldn't matter if it was technically correct or not, the perception would be enough. And frankly, the fact if the FSF is really even considering this casts a bit of a shadow on Linux and Enterprise Support in general: Is it FSF sanctioned businesses only?
Besides, why just Novell? IBM has patent agreements with Microsoft. IBM sells Windows Servers. This seems like nothing more that "We don't like the MS/Novell deal, so let's punish them!"
Okay, actually visiting the site on Vista (RTM release), and the site wants to run MSXML 5.0? Why? I mean, if you are doing a site on how MS stuff sucks, maybe you should have a webpage that doesn't use any of it? Just an idea.
Oh, and just to put my foot in it further, The Donnas are insanely tight live. Straight ahead rock. I saw them before they went major label, it was a small club, and they rocked it nice and hard.
Slashdot Mirror
I am curious as to what super user would actually have 650Gb of stuff sitting on the boot partition.
Been running the RTM for about a week now. And I like it a lot. Sure, I wish Vista users would get this for cheap, but despite that, it really is nice. The new taskbar takes some getting used to, but it has some great features (Having shortcut controls in the preview has a ton of potential). It feels more responsive and polished, even on a netbook that Vista would choke on.
Much fewer hiccups compared to Vista (I can switch around between Visual Studio, Office, Eclipse, etc with no delays). There are lots of surprises. Try the math panel (best with a tablet, but mouse is okay), you can burn ISO images, it has basic color calibration and finally, the taskbar icons have been tamed. Aero Peek is nice and Aero overall takes less resources. It works fine even with Intel integrated graphics.
All in all, it really sets Windows back on good ground. It's the first version of Windows in quite some time that I really like running. Vista wasn't as bad as everybody said, and I'm sure Windows 7 won't be as good as some will hope, but for now, I am happy to run it.
This article was pretty light on real data, but one thing that struck me is that they benchmarked the 32bit version. Well, why you would run the 32bit edition unless you had to, I don't know. I'd like to see some benchmarks with XP/Vista/7 32bit and Vista/7 64bit and see those outcomes.
The numbers for Office show very little difference in performance, so I can't help but feel that the responsiveness gains will make 7 more tolerable to use, and the UI improvements seem to be genuinely interesting and useful. While Vista has been fine for me (64bit version), I expect 7 will be better tuned.
Speaking as a CS postdoc working on a National Library Training fellowship, I have to stay the best bet is to start small and start fresh. There is a ton of praise about VistA, the VA system, but it is built on old storage technology and may not support the kind of analytics needed for quality improvement. Also, there is a lot of information captured in notes, which is free text and not really helpful in terms of decision support and continuity of care.
I think a good place to start would be a couple of things. First, a identity system that links patients at various providers. There wouldn't be one global id, it's would be set of identities tied to a provider. Think a medical OpenID, and that's a start. A ID that is tied to a provider, easy to revoke, and so on.
I think a national medical ID program would be just fine. As long as it stayed optional (you don't have to provide it, but you may need to pay upfront or provide a payment option).
The second is a continuity of care document. A interchange to track the current problem list, current medications, current allergies and list of notable procedures. You don't need everything at once, just the places to ask.
As for security, I think that patients don't mind if their doctors see their record. The system has to be useful first, and making it too secure will make it too hard to use. Also, transparency is a huge issue. You can really build trust if patients can see what is in their record and who accessed it. Health networks really need to have a page for patients to log into and see what's going on.
Also, the legal protections matter, but they have to be useful. HIPPA in its current form has many problems.
Have to say, this is a critical point. While unions have a lot of context with them, they had consistently done one thing well: Improving working conditions for their members. If they make highly repetitive work doable for many, I can't help but imagine that it would help IT a ton.
Developers, I see that as being less of a issue, but for daily operational management, it seems pretty clear the workforce need protection. For one, a union could enable the classic "You are fired for doing nothing" when it is really "I am making sure nothing happens" and so on.
Given the state of broadcast television, I can't say blacking some people out wouldn't do them a favor. Okay, you need to get a convertor box and you may have to wait to get one, but if we encourage people in the meantime to read a book, go to the library, use the computer there and read the news and so on, that's bad? Really?
I mean, I'm scared that people think that TV is that much of a requirement. Local news is nice and all, sure, but you can make do.
"Some vaccines are unneeded. Chicken pox, for example..."
Well, until there is science noting that the risks of the vaccine are higher than the risks of chicken pox infections (which can lead to serious complications and deaths in children and adults), there are good reason to vaccinate. The severity of the disease is not the issue. The issue is the vaccine simulating an effective immune response. The CDC guidelines are taken very, very seriously. So, again, look at the science they do before deciding what is needed or not.
But I'd bet the small business owner is a hell of a lot happier following their dreams on their terms. The crazy thing about this society is that we need people to make our sandwiches, cut our meat and so on, but take pride in belitting them. Not everybody can run a business. No everybody has the chances others get. Others may care about other things. There are other ways to contribute to society that aren't driven by the bottom line. Most SBOs I know just love what they do and enjoy it. But, you know what? It's not about the time you put in. Working eighty hours on your business can mean you are super motivated, or you don't have anything else to do.
The small business owner has a chance for promotion, for success, to make a difference, for all that hard work to make a difference. Too many in this country work hard to survive, not thrive. They work 50 hours or more to get by. The working class may not be "the hardest working", but they are the ones who would hardest for the least in return. Maybe things would be a bit better if the SBOs were more humble and took pride in provide a living for others and less in how much they cleared this year.
I know, offtopic, but are their cheap scanners out there now? I'd like to inventory my CD, books, etc.
Firstly, I feel that people are often looking at scooters and low-displacement motorbikes. For urban drivers, we are talking 100 mpg from a vehicle that is in the 2000-3000K range. Oh, and don't forget the people on bicycles. on
Yes, bikes are more risky, but those risks are:
The driver. But, stupid drivers can get hurt in any vehicle.
Cars. I got wrecked because a car decided it was okay to head into my lane. The bike stood up, I ran out of road and that's that. The guy didn't even notice.
As more and more people get on two wheels, I just ask this of car drivers:
Pay attention. Get off the phone. Don't fuss around.
Start Seeing Motorcycles. Look for us. We are there. We have a right to be on the road too, and a lot of us are really okay.
As for motorcyclists, our part of the bargain:
Know your limits, practice your skills and wear your gear.
Take it easy. If you need a race, find a track.
Harley is studied everywhere in marketing. I mean, how many companies have people tattoo their logo on themselves. Wow.
Yes, there are Harley's with fuel injection. This happened a few years ago, and yes, they have a engine that doesn't totally suck. But their bikes are very costly for what you get. I mean, the cheapest is about 8400. Hint. Do not buy that bike. There are way, way better bikes for 2000-3000 less. Hell, you get more for a BMW at that price.
Scanning through the highly moderated article, I didn't see this point raised, which seems critical. The doctors that would be required to do this research to provide doping and drugs are bound by a ethical code of conduct, and this definitely forbids giving people medications with very serious risks to somebody that is healthy.
Some have expression the opinion of "let them be lab rats." Well, I doubt that any good doctors would touch that ethical quagmire with a ten foot pole. I mean, the notion of informed consent alone is a huge red flag. Can you say that somebody that is willing to risk their health and life to potentially (nothing is certain) perform a bit better is truly of a state of mind to consent? And what doctor would do such a thing. Would you want a doctor that doesn't care about your health, but just want they can get away with?
And the ethics are there for a reason. Bioethics exists for a reason. Part of the Human Genome Project was funding to examine carefully and intensively the impacts of the project, how the data should be used, the impacts on society. In fact, there was discuss if the project should be allowed at all.
Of course, things can improve in testing. What I think needs to happen (and has, in the NFL, for example), is that more athletes need to come out and say that they want to test their limits without risking their health. That at the end of the day, they don't want to dope. That they are not willing to do anything to win. They want to test themselves on their terms, not as some "mad scientist" experimental rat.
And the sport fan needs to learn that the best part of sports is not winning or losing, it is the pursuit of excellence and being tested and retested time and time again.
Corticosteroids (the kind used to treat psoriasis) are a completely different animal from anabolic steroids when it comes to human performance. Yes, they share some chemical commonalities, but croticosteroids don't raise testoterone levels (anabolic steroids do). Also, an barbiturates probably aren't considered performance enhancing either.
Of course, all of the above have side effects (I'm sorry your dad doesn't respond to other anti-seizure meds, the barbs suck). But the reasons for this is that those effects are not as bad as the the problems they address.
When I read this, it was more evidence that some in Google want to be "the internet". For very, very many, they are the gatekeepers to the web. Now, with this, they can be the hosts as well. If Google hosts a ton of applications, you can see a future in which somebody is surfing from site to site but never really leaving Google.
So, what's to stop the extingush part? Letting Google say, hey, host here and we will make sure that your search results don't "change", or you get better results, etc. Given the demands on growth and the need for Google to justify it's P/E, it is starting to look for more ways to make money. And Google is large enough to become the "Microsoft" or "IBM" of web applications if it really wants too, and why wouldn't it?
Of course, I don't buy the "Google isn't evil, that's why" prima facie argument.
Yea, it's going to be tricky. Big issue. For the longest time, we got a huge boost from Moore's Law in single-threaded performance, and for desktop OSes (esp. XP, Vista), they are tuned and around interactive, single foreground program workloads. But that's over. And Vista does add IO and memory demands (search, indexing, UI experience, etc.), and some in the market (and MS, of course) want that.
Add in memory hierarchies from hell (L3, it's back, it's crazy) and it does get to the point where you have to think about stuffing as much stuff as you might need into memory, because going to disk just kills you, because going to ram to cache is bad enough.
Well, just to be clear, I just have 4G right now on my workstation and it's more than fine. Even 2G is no problem (older laptop). But, yes, I stand by my statement that for my work, I'd rather have lots of slower memory than tons of fast memory. YMMV of course.
I've been running Vista for about 16 months now. (Got it from MSDN, etc.). I ran both 32bit and 64bit editions, on two kinds of hardware. It's been a bumpy ride sometimes.
If at all possible, use the x64 edition. Yes, some OEM make this a pain, but try. Given this, my next suggestion seems obvious. Get more memory. The more, the better. I'd rather have 8G of DDR2-533 than 2G of DDR3 uber-awesome overclocked OMGBBQ ram, because caching works. If on an Intel integrated graphics, turn down some of the Aero stuff. Duh. If possible, just buy a cheap 8400GS, because even that will help.
SP1 helps. Some things are faster. Of course, I'm not seeing some of the problems others are. I recently shipped some ISOs over from machine to my server (Win 2008), and it just flew. Got about 600Mb over a 1Gb switched link.
All in all. Not one blue screen on the desktop, a couple on the laptop due a older bluetooth driver. Things seem plenty responsive and fast, but there is a breaking in period. Sure, it isn't "awesome", but ME it sure is not.
Be patient! Indexing and the prefetching stuff takes time, but it does work. I use Outlook (okay, I know, I know) a lot, and it fired up faster and faster after the first day or two.
Nothing official, but I hear that the MinWin idea was based a lot on the Singularity project. The link has some papers, and there is some really interesting ideas there in terms of fast process isolation, etc. Sure, a microkernel is nothing new, but there are new ways of making a microkernel.
Or Microsoft. Or IBM. And doesn't RedHat count with the whole JBoss thing?
I'd be thrilled to have a motorcycle jacket and armour made better by this stuff.
As for some of the decisions, they are not just about pure speed. There are more considerations.
Value types: This allows the choice of the best semantic match is. Also, this does help with native interoperability, as in some cases, you can model a C struct in C# and pass it back and forth with a lower cost. Also, the complexity cost you note is small, as even Java still has value types. If you have to deal with both, then adding in structs isn't that costly, from a language and runtime view.
JIT-only: The ability to pre-compile to native code helps with native interoperability and performance. It can affect startup time, but there are techniques to deal with this a bit. I think MS thought the added complexity of interpretation+hotspot compliation wasn't worth it for them. It was a tradeoff, but I don't see it as a drastic one. Also, C# can still code-pitch if needed. It is probably harder.
Real Generics: To be honest, I can't say that Java Generics are all that great, from a programming model. Sure, they don't technically cost anything, because they go away. Which causes real problems and limitations. For example, C# 3.0 can do a ton of type-inference and more functional-like program as the type information of a delegate (or lambda expression) is kept and passed. Also, I'd like to see the reference for the CLR only inlining one deep.
Embedded Native Code: Well, faster access to native code can make a huge improvement in performance, and this is a real problem with the Java runtime. JNI has some real issues in terms of usability from a programming standpoint. The CLR makes it must easier to call into a native function if you need to, and that's often a really useful point. Not just for Windows, but for OS/X, for Linux, etc.
The GC: Sure, it is more complex, but there isn't much evidence that it is slower because of it. Again, if fast native interoperability is a goal, then this may be a cost, but it may be a good one.
Yes, the CLR is more complex. That is because it can actually model more aspects of an executable. It has pointers. It has "unsafe" code. You can compile any C/C++ program that compiles in Visual Studio 2005/2008 to the CLR with a few switches. In some cases, the performance hit is surprisingly small versus the native code. With the Phoenix project, the .Net CLR has access to the same optimizations that the native C/C++ compiler has.
All in all, the CLR, C# do represent some interesting ideas and is a very valid platform. The MS equals sucks argument just doesn't hold here. But, it's just an opinion.
I gone through the initial process with Google twice, with the same outcome. It seems to me they need to improve their HR process, as I've gone through a phone interview, but then told I wasn't a good fit.
If you look at my background and resume, I think you would concur that the positions I was interviewed for weren't a good fit, but because it was Google, I gave it a shot. But, fool me once, fool me twice and all that. If they call again, I'll let them know how I feel about the whole process.
If the FSF actually tried to do this, it'd be great for Microsoft. It would basically allow Microsoft to say that Linux can't be taken seriously by any business, because the FSF will basically revoke your "right to use" Linux if they don't like how you do business (do you have software patents? No Linux for you).
It wouldn't matter if it was technically correct or not, the perception would be enough. And frankly, the fact if the FSF is really even considering this casts a bit of a shadow on Linux and Enterprise Support in general: Is it FSF sanctioned businesses only?
Besides, why just Novell? IBM has patent agreements with Microsoft. IBM sells Windows Servers. This seems like nothing more that "We don't like the MS/Novell deal, so let's punish them!"
Okay, actually visiting the site on Vista (RTM release), and the site wants to run MSXML 5.0? Why? I mean, if you are doing a site on how MS stuff sucks, maybe you should have a webpage that doesn't use any of it? Just an idea.
christ girls, there's more to the guitar than power chords
No, really?
Oh, and just to put my foot in it further, The Donnas are insanely tight live. Straight ahead rock. I saw them before they went major label, it was a small club, and they rocked it nice and hard.